On 09/03/10 12:09, Dave Abrahams wrote:
Can someone explain to me, in English, what this technique allows me to express that I couldn't otherwise?
Is this just about allowing generic composition of lambdas?
Nothing, AFAICT. What it does do is make implementation of beta reduction easier to implement because one doesn't have to do any alpha conversions. I think one has to do some decrementing of the NestLevel in the arg<NestLevel,ArgIndex> terms, as suggested by: 2. decrease the free variables of M to match the removal of the outer λ-binder From: http://en.wikipedia.org/wiki/De_Bruijn_index Now I'm guessing here David, but when I've encountered problems with mpl's apply<F,A1,A2,...,An>, sometimes I've had to wrap F in an mpl::lambda to get it to work Now one thing that this DeBruijn technique *might* do, is simplify that lambda, because mpl::lambda looks real complicated to me. Maybe Mr. Sankel could give that a try, although his code only deals with values and not types. For those unfamiliar with the beta and alpha terms, this may help: http://en.wikipedia.org/wiki/%CE%91_conversion#.CE.B1-conversion